Mathematical modeling of vibromyographic signals from skeletal muscle

n the first study, we adopt the vibrating string and bar theories to model the muscle vibration in a twitch. The equation that governs the muscle vibration is derived with some simplification. The numerical simulations are performed with different mechanical and physical properties to simulate how those properties influence on the characteristics of VMG signals. The following results have been obtained: (1) The frequency and amplitude of the VMG signal are modulated by both muscle tension and muscle length. Larger tension and shorter length cause VMG signal with larger amplitude and higher frequency; vice versa. (2) The viscosity coefficient can only affect the amplitude of the VMG signal. The larger viscosity coefficient causes the VMG signal with lower amplitude. (3) The initial condition (the initial deflection of the muscle) affects the amplitude of the VMG signal greatly. The larger the initial deflection, the larger the amplitude will be. (4) The stiffness, which is an important indicator of mechanical and physiological properties of muscle, can be well modeled as a fibrous cable rather than solid bar.;In the third study, a mathematical model for the generation of VMG signal under voluntary contraction is proposed by using multichannel model, which has been widely utilized in EMG signal modeling. The characteristics of power spectral density (PSD) of the VMG signal is discussed. It is found that the PSD of VMG signal is much more sensitive to firing statistics parameter (mean firing rate) than that of EMG signal, suggesting that the VMG signal has advantage over EMG signal when indicating central nervous system fatigue.;In the second study, the time-frequency method is used experimentally and theoretically to estimate the mean and the variance of the frequency of VMG signal during an electrically elicited twitch. Some features of muscle fatigue have been observed as follows: (1) The root of mean squared (RMS) values of the amplitude of VMG signal may not be an adequate indicator of fatigue. (2) The parameter...